Quantifying 3D tracer velocity and porosity on core scale from 3D GeoPET image sequences


Quantifying 3D tracer velocity and porosity on core scale from 3D GeoPET image sequences

Eichelbaum, S.; Lippmann-Pipke, J.; Korn, N.; Kulenkampff, J.

Flow and transport simulations in geomaterials are commonly conducted on high-resolution tomograms (µCT) of the pore structure or stochastical models that are calibrated with measured integral quantities, like break through curves (BTC). Yet, there existed virtually no method for experimental verification of the simulated velocity distribution results.
Positron emission tomography (PET) has unrivaled sensitivity and robustness for non-destructive quantitative spatio-temporal measurement of tracer concentrations in body tissue. We empowered PET for its applicability in opaque/geological media (GeoPET). Thus it is the appropriate method for experimental verification and calibration of computer simulations of pore-scale transport by means of the observed propagation of a tracer pulse, cPET(x,y,z,t).
As a principal concept, velocity and residence time distributions, as well as the tortuous pathway topology, principally can be derived directly from cPET(x,y,z,t). However, the fundamental experimental limit of finite signal to noise ratios is manifested in apparently intermittent propagation pathways, with gaps in zones where the concentration cPET falls below the significance threshold. This hampers the direct parameter estimation of velocity, v(x,y,z), and porosity distribution n(x,y,z) from cPET(x,y,z,t).

This issue is overcome to some extent by introduction of causality (continuity) into the here presented evaluation algorithm.

Likely Topics in math & comp. sciences
1) Space-Time Processes
2) Image Analysis
3) Numerical Modelling and Numerical Simulation
4) Inverse Problem solving
5) other computer sciences methods

Likely Geoscience topics:

1) Water: sea, surface and subsurface

  • Lecture (Conference)
    IAMG 2015, The 17th annual conference of the International Association for Mathematical Geosciences, 05.-13.09.2015, Freiberg, Deutschland

Permalink: https://www.hzdr.de/publications/Publ-21892
Publ.-Id: 21892